Live viral vaccines have traditionally offered the most effective protection against viral infections. Such vaccines induce strong cellular and humoral immune responses due to high level intracellular synthesis of antigens over extended periods. Live attenuated vesicular stomatitis virus (VSV) vectors expressing proteins of other viruses are highly effective vaccines in animal models. VSV vectors expressing HIV-1 Env and SIV Gag proteins have proven effective in protecting against AIDS in a rhesus macaque model. The first generation of VSV vectors used in macaques expressed SHIV gag and env genes from a downstream position in the VSV genome. Although these vectors expressed substantial amounts of SHIV proteins, new vectors have now been developed that allow even greater protein expression from an expression site in the first position in the VSV genome. Proteins expressed from the first position are the major proteins produced in infected cells. The major goal of this project is to test the effectiveness of these new vectors in rhesus macaques and to include expression of Pol protein to generate immune responses to additional CTL epitopes. The magnitude of the immune responses will be followed quantitatively and compared with previous studies, and the extent of protection from challenge will be monitored through analysis of peak viral loads, CD4 T cell counts, viral load set points, and protection from AIDS. The optimized VSV vector expression system will also be used in a second study to determine if the VSV-based vaccine system can also protect from AIDS in a challenge using SIV rather than SHIV89.6p. This later series of experiments is important because SIV infection of macaques more closely mimics HIV infection in humans. The proposed studies will also continue the follow up on the seven SHIV-infected, vaccine-protected macaques from the initial grant period to determine durability of vaccine protection.